Mycobacterium tuberculosis remains the bacterial leading cause of death worldwide and multidrug-resistant strains have emerged as a new problem in both developed and developing countries. The cell wall of M. tuberculosis has a unique, complex structure and is extraordinarily thick, rigid, and hydrophobic. Because of these characteristics, it is highly impermeable to ordinary antimicrobial agents. Bacterial autolysins are enzymes which are capable of hydrolyzing the bacterial cell wall and are associated with normal bacterial cell division, growth, and autolysis. Little is known about the autolysins of mycobacteria. The central hypothesis of this proposal is that mycobacterial cell wall autolysins are essential for the growth and survival of the organisms and can be exploited as new targets for anti-mycobacterial agents. In preliminary studies, the investigators have (i) prepared a mycobacterial cell lysate that hydrolyzed the cell wall polysaccharide; (ii) observed that ethambutol treated mycobacteria have increased cell wall hydrolysis; and (iii) identified an open reading frame (ORF) in the M. tuberculosis genome that is highly homologous to a Bacillus subtilis cell wall hydrolases, i.e., an N-acetylmuranoyl-L-alanine amidase. The investigators now propose the following: AIM1. The putative M. tuberculosis amidase gene noted above will be amplified by PCR, cloned and expressed. If the protein is confirmed to have the predicted enzymatic activity, then its functional and biochemical characteristics will be determined and the expression properties of the gene will be analyzed. AIM 2. (a) Develop rapid and sensitive in situ phenotypic assays and apply these to screening for cloned mycobacterial autolysins expressed in E. coli. (b) Develop improved zymography and cell-free enzymatic assays and apply these to the isolation and purification of putative hydrolases. Purified enzymes will be analyzed for structural information (e.g. N-terminal sequence, quantitative protein mass as determined by MALDI/TOF mass spectroscopy), which will be used to identify the encoding gene in the published M. tuberculosis genome sequences.